MLX TTS for Openclaw

A high-performance local text-to-speech engine for macOS that leverages Apple Silicon and MLX for fast, private audio synthesis.

guoqiao
v0.0.3
Feb 10, 2026
0
2.5k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install mlx-tts

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install mlx-tts using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is MLX TTS?

MLX TTS is a specialized tool within the Openclaw Skills ecosystem designed for high-performance audio synthesis exclusively on macOS. By leveraging the Apple Silicon MLX framework and state-of-the-art open-source models such as QWen3-TTS, it provides users with a free, fast, and serverless way to convert text into speech.

This integration ensures that all audio processing remains strictly on-device, offering superior privacy and zero reliance on external API keys. This capability makes it an essential asset for developers looking to build responsive, voice-enabled agents using Openclaw Skills that function entirely offline without sacrificing speed.

MLX TTS Use Cases

  • Convert text prompts or long-form documents into audible voice files locally on macOS.
  • Enable AI agents to respond to user queries with high-quality, synthesized voice messages.
  • Create accessibility-focused workflows that narrate text without sending sensitive data to the cloud.
  • Generate audio assets for media projects or presentations using local open-source models.

How MLX TTS Works

  1. The user provides a text prompt via a command like /mlx-tts or a natural language request.
  2. The Openclaw Skills agent executes the mlx-tts.sh shell script, passing the text as an argument.
  3. The MLX framework processes the text through the QWen3-TTS model locally on the M-series processor.
  4. The script generates an audio file (typically in .ogg format) and returns the absolute file path to the agent.
  5. The agent utilizes its internal messaging tool to deliver the generated audio file to the user as a voice message.

MLX TTS Setup

To get started with this entry in the Openclaw Skills library, ensure you are on macOS with Apple Silicon and have Homebrew installed. Run the following installation script:

bash ${baseDir}/install.sh

This script automatically handles dependencies, installing uv for Python package management and mlx_audio for the core synthesis tasks. To manually generate audio, you can use:

bash ${baseDir}/mlx-tts.sh "Your text here"

MLX TTS Data Schema & Taxonomy

The skill manages data through local temporary directories and standard CLI outputs.

Data Component Details
Input Type Plain text string passed via CLI argument
Output Type Local file system path to the generated audio file
Audio Format Default output is .ogg (compatible with Openclaw Skills messaging)
Environment Managed via uv to ensure isolated Python dependencies

MLX TTS Advanced Features

  • Native Apple Silicon optimization for near-instant inference speeds using the MLX library.
  • Privacy-first architecture with 100% local execution and no external API dependencies.
  • Automated environment setup using uv to handle complex Python dependencies seamlessly.
  • Support for the high-quality QWen3-TTS open-source model out of the box.
  • Seamless integration with Openclaw Skills messaging tools for automated agent-to-user voice responses.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Bins brew
Github Stars: 0
forks: 0

Featured*